Computed tomography-based radiomics diagnostic approach for differential diagnosis between early- and late-stage pancreatic ductal adenocarcinoma

Table 1 Demographic characteristics of patients with early-stage and late-stage pancreatic ductal adenocarcinoma

Characteristic	Clinical stage classification		P value
	Early-stage PDAC (n = 31)	Late-stage PDAC (n = 40)
Age (yr)	61.45 ± 8.93	61.63 ± 8.31	0.933
Sex			0.153
Male	20 (64.5%)	19 (47.5%)
Female	11 (35.5%)	21 (52.5%)
Abdominal pain	14 (45.2%)	20 (50%)	0.686
Abdominal bloating	8 (25.8%)	11 (27.5)	0.873
Abdominal discomfort	16 (51.6%)	21 (52.5%)	0.941
Body weight loss	11 (35.5%)	14 (35.0%)	0.966
Jaundice	7 (22.6%)	9 (22.5)	0.994
AJCC staging			-
IA	4 (12.9%)
IB	8 (25.8%)
IIA	3 (9.7%)
IIB	16 (51.6%)
III	0	16 (40%)
IV	0	24 (60%)

PDAC: Pancreatic ductal adenocarcinoma; AJCC: American Joint Committee on Cancer.

Table 2 Computed tomography findings of patients with early-stage and late-stage pancreatic ductal adenocarcinoma

Characteristic	Clinical stage classification		P value
	Early-stage PDAC (n = 31)	Late-stage PDAC (n = 40)
Tumor size (cm)	4.03 ± 1.25	3.05 ± 0.54	< 0.001
Tumor location			0.869
Head & neck	18 (58.1%)	24 (60.0%)
Body & tail	13 (41.9%)	16 (40.0%)
Tumor margin			0.722
Well-defined	4 (12.9%)	3 (7.5%)
Ill-defined	27 (87.1%)	37 (92.5%)
Calcification			1.0
Y	1 (3.2%)	1 (2.5%)
N	30 (96.8%)	39 (97.5%)
Cystic degeneration			0.757
Y	3 (9.7%)	6 (15.0%)
N	28 (90.3%)	34 (85.0%)
MPD			0.861
Y	20 (64.5)	25 (62.5%)
N	11 (35.5)	15 (37.5%)
Vascular invasion			0.302
Y	14 (45.2%)	23 (57.5%)
N	17 (54.8%)	17 (42.5%)
Lymph node metastasis			0.480
Y	16 (51.6%)	24 (60.0%)
N	15 (48.4%)	16 (40.0%)
Distal metastasis			< 0.001
Y	0 (0%)	24 (60%)
N	31 (100%)	16 (40%)

Categorical variables were presented as the number of cases (%) and were analyzed by using χ² or Fisher’s exact tests. MPD: Main pancreatic duct dilatation; PDAC: Pancreatic ductal adenocarcinoma.

Table 3 The differential diagnostic ability of the selected radiomics features

Features	AUC	SEN	SPE	ACC
a_Compactness2	0.754	67.5%	77.42%	71.83%
a_GLCMEntropy_AllDirection_offset7_SD	0.698	82.5%	58.06%	71.83%
p_Correlation_angle90_offset7	0.683	60.0%	70.97%	64.79%
p_HighGreyLevelRunEmphasis_AllDirection_offset4_SD	0.661	77.5%	51.61%	66.20%
a_GLCMEntropy_AllDirection_offset4_SD	0.665	87.5%	54.84%	73.24%
a_Correlation_angle45_offset7	0.639	50.0%	77.42%	61.97%
p_Compactness2	0.66	80.0%	54.84%	69.01%
a_InverseDifferenceMoment_AllDirection_offset4_SD	0.638	75.0%	64.52%	70.42%
a_Correlation_angle135_offset4	0.677	75.0%	58.06%	67.61%

AUC: Area under the receiver operating characteristic curve; SEN: Sensitivity; SPE: Specificity; ACC: Accuracy; GLCM: Grey level co-occurrence matrix.